The structure and properties of viscoelastic polyurethane
foams with fillers from coffee grounds*
)
Monika Auguścik-Królikowska1), **), Joanna Ryszkowska1), Agata Ambroziak1),
Leonard Szczepkowski2), Rafał Oliwa3), Mariusz Oleksy3) DOI: dx.doi.org/10.14314/polimery.2020.10.6
Abstract: The subject of the research presented in this article are viscoelastic polyurethane foams (VPF)
made using a fillers from coffee grounds. The foams were made with varying content of coffee fillers. Foams and fillers were characterized by means of techniques such as infrared spectroscopy, thermo-gravimetric analysis, differential scanning calorimetry and cone calorimeter. The changes in the mi-crostructure of the foams were analyzed using a scanning electron microscope. Foam properties in the compression test was assessed. As a result, it was concluded that the use filler from coffee grounds causes a significant reduction in compression set of foams after compression by 75% and by 90% of the height of the samples. It was shown that these changes are caused by changes in the chemical structure of the macromolecules of these foams but also in their structure. After addition the filler, all foams have permanent deformations with a value below the permissible limit of 10%. Addition of 20% by mass of filler caused by a decrease in the rate of heat release and the rate of smoke release during foam combus-tion.
Keywords: viscoelastic polyurethane foams, waste of coffee, grounds, structure, permanent
deforma-tion, hardness.
Struktura i właściwości lepkosprężystych pianek poliuretanowych
napełnionych fusami kawy
Streszczenie: Przedstawiono wyniki badań lepkosprężystych pianek poliuretanowych (VPF)
wytwo-rzonych z dodatkiem różnych ilości fusów kawy. Pianki i napełniacz scharakteryzowano z zastoso-waniem spektroskopii w podczerwieni, analizy termograwimetrycznej, różnicowej kalorymetrii ska-ningowej i kalorymetru stożkowego. Zmiany w strukturze pianek obserwowano metodą skaska-ningowej mikroskopii elektronowej. Oceniono właściwości pianek w próbie ściskania. Stwierdzono, że napeł-niacz w postaci fusów kawy powoduje zmniejszenie odkształcenia trwałego pianek w próbie ściskania o 75% i zmniejszenie o 90% wysokości próbek, co jest wynikiem zmian w budowie makrocząsteczek pianek i w ich strukturze. Wprowadzenie 20% mas. napełniacza spowodowało zmniejszenie szybkości wydzielania ciepła i dymu w trakcie spalania pianek.
Słowa kluczowe: lepkosprężyste pianki poliuretanowe, odpady kawy, fusy, struktura, odkształcenie
trwałe, twardość.
Coffee is a very popular drink, which makes it the
sec-ond largest sold commodity in the world, secsec-ond only to crude oil [1, 2]. According to the International Coffee Organization 9.5 million tons of coffee is produced in
2018, and its production is steadily increasing.
Spent coffee ground (SCG) is the residue obtained dur-ing the brewdur-ing process [3]. The huge amount of residue arise too in the production of soluble coffee. Annually, around 6 million tons of coffee grounds are produced all over the world.
Currently, SCGs are used for composting, gardening, bioenergy production, mushroom growth. By 2020, more coffee waste is expected to be used for energy production [2]. Lack of proper management of SCGs can lead to envi-ronmental pollution [4]. That is why a lot of research is car-ried out to develop them for example: for biodiesel
produc-1) Warsaw University of Technology, Faculty of Materials
Science, Wołoska 141, 02-507 Warszawa, Poland.
2) FAMPUR Adam Przekurat Company, Gersona 40/30, 85-305
Bydgoszcz, Poland.
3) Rzeszow University of Technology, Faculty of Chemistry,
Department of Polymer Composites, al. Powstańców Warsza-wy 6, 35-959 Rzeszów, Poland.
*) Material contained in this article was presented at the
Science and Technology Conference on “Polyurethanes 2019”, Ustroń, Poland, 13–16 October 2019.
tion [5, 6], as source of sugars [7] and as sorbent for metal ions removal [8, 9]. Coffee beans contain large amounts of organic compounds (i.e. fatty acids, lignin, cellulose, hemi-cellulose, and other polysaccharides). This includes a sig-nificant amount of carbohydrates, minerals and proteins, as well as caffeine, tannins and free phenols [2, 10, 11]. Some of these compounds are toxic, which limits the pos-sibility of using SCGs as animal feed. In addition, SCGs contain a high moisture content, which significantly hin-ders their management, due to putrefaction processes [12] The chemical structure and composition of coffee waste particles is similar to the structure of its beans [13], they are rich in carbohydrates (forming primary structures, i.e. hemicellulose, cellulose, lignin) and so-called secondary structures, i.e. pectins, proteins and bioactive compounds [2]. Literature review revealed that about 40–50 wt % of SCG is polysaccharides as cellulose and hemicellulose structures [14–17]. Spent coffee ground (SCG) contains: fatty acids, amino acids, polyphenols, minerals and poly-saccharides that can be exploited as a source of value-added products [2]. The SCG were characterized for their moisture content (12.1 wt %), cellulose (33.6 wt %) and total lignin (13.8 wt %) [5].
Various groups of natural fillers: mineral, animal and vegetable origin can be used to produce polymer com-posites from thermoplastics and duroplasts [18]. In the group of fillers of plant origin, different parts of plants can be used: bast, leaves, seeds, stems, grasses, reeds, husks and shells. Coffee grounds may be included in this group of fillers. When using plant fillers for the pro-duction of polyurethanes, it is important that the hydro-philic cellulose contains three hydroxyl groups (OH) eas-ily absorbs moisture, it can also contain carboxyl groups (COOH), the other component hemicellulose contains both OH and COOH groups. Both components contain functional groups that can react with isocyanate groups. On the other hand, secondary ingredients may e.g. act as plasticizers. The subject of the article are composites of viscoelastic polyurethane foams modified with a filler obtained from coffee grounds.
Viscoelastic polyurethane foams (VPF), also known as shape memory foams, is characterized by slow recovery after compression [19]. The main application is the high quality mattress and pillow segment, where viscoelastic foams distribute the body weight over a larger area. For several years these foams have been of particular interest in the industry [20, 21].
EXPERIMENTAL PART Materials
The matrix of the composite was viscoelastic foam (VPF), (know how – FAMPUR Adam Przekurat). Foams were produced with the isocyanate index – INCO = 90. The following components were used to make the mate-rials:
– polyol Daltocel F442, F526 Huntsman Polyurethanes; – polyol Rokopol F3600 z PCC Rokita SA, Poland; – polyol Rokopol M1170 z PCC Rokita SA, Poland; – isocyanate Ongronat 4040 BorsodChem, Hungary; – foaming agents, water;
– catalysts; – surfactants.
As modifiers were used filler from spent coffee grounds – post-consumer grounds of coffee obtained from the Costa Cafe in Warsaw. The dried waste was milled using a coffee grinder. The foams were produced by a one-step method. Polyol component A containing polyols, catalysts, surfactants, water, catalysts, and filler were prepared for the composites. Then, using a stirrer at 3000 rpm, component A was mixed for 10 s, then com-ponent A was mixed with comcom-ponent B (isocyanate) for 10 s and poured into an open rectangular mold. During the synthesis, the start time and foam growth time were recorded. After the synthesis, the foam was heated at 70°C for an hour, then seasoned for 7 days at room temperature. Only after this time, test samples were cut. The sample names, the foams’ composition are given in Table 1. T a b l e 1. Description of tested materials
Sample VPF0 VPF10 VPF20 VPF30 VPF40 Filler content, php – parts per
hundred polyol by weight 0 10 20 30 40 Methods of testing
– The apparent density (d) was calculated by measur-ing the mass and volume of the sample. The mass of the samples was determined with an accuracy of ±0.001 g, and the dimensions of the samples were measured with an accuracy of ±0.01 mm. The apparent density of foams was determined according to PN-EN ISO 845:2010 standard.
– The chemical composition was analyzed using a Nicolet 6700 (Thermo Electron Corporation) – FT-IR spectrophotometer equipped with an ATR unit (sup-pressed total reflection). Each sample was scanned
64 times in the 400–4000 cm-1 wavelength range,
absorp-tion spectra were recorded. The results were analyzed using OMNIC Spectra 2.0 software (Thermo Nicolet).
– Thermogravimetric analysis was performed using a TGA Q500 TA Instrument under a nitrogen atmo-sphere. Samples of approximately 10 mg were heated in platinum dishes from room temperature to 700°C, with a heating rate of 10°C/min.
– Differential scanning calorimetry was performed using a DSC Q1000 from TA Instruments. Measurements were made in a helium atmosphere in hermetic aluminum crucibles. Samples of approximately 6 mg were heated from -80°C to 200°C with a heating rate of 10°C/min.
– A Hitachi SEM TM3000 scanning electron micro-scope was used to observe the microstructure of the produced materials. Before observation, filler and foam
samples were sprayed with a palladium gold layer. Observations were made using a 5 keV voltage.
– Compressive set (CS) deformation was determined on
50 × 50 × 50 (mm3) samples in the direction of foam growth.
Foams were compressed by 50%, 75% and 90% of the origi-nal height in metal covers. Then they were placed in an oven at 70°C for 22 hours. After this time, the samples were removed from the metal covers and after 0.5 hours at room temperature their height was again measured. The percentage loss in height of samples was calculated, thus determining compression set: CS (22 h, 70°C, 50%), CS (22 h, 70°C, 75%), CS (22 h, 70°C, 90%). The test was carried out according to the standard PN-EN ISO 1856:2018-09.
– The foam compression process on the Zwick Z005 testing machine was also assessed. The samples were com-pressed by 75% of their height. The value of compressive stresses during loading and unloading was obtained, thus obtaining a hysteresis loop. Each sample was squeezed 4 times. The comfort coefficient (SAG factor) of the foam (stress at 65% compression of the sample/stress at 25% height of the sample) and the hardness of the foam (stress at 40% height of the sample) was determined on the basis of the compression test. The test was carried out according to the standard PN-EN ISO 3386-1:2000.
– Particle size was determined in a Horiba LA-950 laser particle size analyzer using the LALLS (Low Angle Laser Light Scattering) technique to achieve full geometric and morphological characteristics of powders. The measure-ments were carried out with the refractive index specific to cellulose (1.47) laser setting, which reports the full geo-metrical and morphological characteristics of powders.
– The foams were also characterized using a cone cal-orimeter (Fire Testing Technology Ltd.) in accordance with ISO 5660:2002. The samples with the dimensions of
100 × 100 × 8 (mm3) were tested by heat flux at 35 kW/m2.
RESULTS AND DISCUSSION Filler characteristics
The structure of the filler (SCG) is similar to other plant fillers, which are composed of three main primary
com-ponents: cellulose, lignin and hemicellulose, and sec-ondary components, i.e. extracts, waxes and others [5]. To describe the structure of the SCGs used, thermogravi-metric analysis (TGA) and analysis of the derivative of mass changes as a function of temperature (DTG), dif-ferential scanning calorimetry (DSC), chemical structure analysis using Fourier transform infrared spectroscopy (FT-IR) and scanning microscopic observations electron microscopy (SEM). The results of thermogravimetric analysis of SCG are shown in Fig. 1.
VIII degradation stages were observed on the DTG
curve. At each stage, the mass loss at a given stage (Δmi)
was estimated. For selected degradation stages, the tem-perature of the maximum degradation rate in a given
stage (Tmax) and the degradation rate (vmax) were
deter-mined on the DTG curve. The results of this analysis are summarized in Table 2.
The SCG were dried but total elimination of water was difficult due to the hydrophilic nature of the particles, which is present even as structurally bound water mol-ecules. The first one step up to 120°C representing the mass loss in particles could be related to water loss asso-ciated with moisture present in the particles [22–24]. The initial stage of the SCGs degradation was associated with loss of about 3.6 wt % water. Compared to other natural fibers, which may contain up to approx. 15 wt % is a small T a b l e 2. Results of TGA and DTG analysis of SCG in subsequent steps of degradation
The SCG
degradation stage ΔT°Ci wt %Δmi T°Cmax wt %/°Cνmax Description of the SCG degradation stage
I ≤120 3.6 ± 0.5 55 0.05 Water
II 120–200 1.1 ± 0.5 175 0.04 Lipids, proteins, caffeine, tannins or free phenols
III 200–340 42.7 ± 0.9 314 0.63 Hemicellulose
IV 340–390 15.5 ± 0.7 345 0.48 Cellulose
V 390–445 6.5 ± 0.5 410 0.14 Lignin
VI 445–500 4.7 ± 0.5 450 0.09 Lignin
VII 500–600 6.9 ± 0.5 475 0.08 Lignin and aromatic ring
VIII ≥600 19.5 ± 1.1 Ash 100 80 60 40 20 0 Mass, % 0.8 0.6 0.4 0.2 0.0 Deriv ativ e mass, %/°C 50 150 250 350 450 550 650 750 850 950 TmaxIII TmaxIV III IV Tma x I Tma x II I II V VI VII VIII TmaxV Temperature, °C
amount of water [25]. The maximum degradation rate in this step is reached at 55°C. The moisture content of the fibers dependent on the content of non-crystalline parts of cellulose and the void content of the particles [26, 27].
Therefore, it can be assumed that during the second stage degraded, bound water in the structure and the so-called second order ingredients: proteins, caffeine, tan-nins or free phenols. During this stage, about 1.1% of the sample mass was lost, and the maximum degradation rate at this stage is lower than during the first stage and was achieved at 175°C.
The subsequent stages of SCG degradation are associ-ated with the degradation of the primary components of vegetable fillers: hemicellulose (200–320°C), cellulose (250–380°C) and lignin (280–500°C) [28, 29]. During stage III, hemicellulose is degraded, thermal stability of hemi-cellulose is weaker than that of hemi-cellulose, which is attrib-utable to easy ring opening, ring cleavage, dehydration or deacetylation reproducing monosaccharides. The maxi-mum degradation rate at this stage is the highest among all degradation stages and is achieved at a temperature of approx. 315°C. In the next stage, cellulose decomposes.
Cellulose is ordered and with a long polymer of glu-cose with a glycosidic bond without branches, although it also turns into low molecular hydrocarbons owing to thermal decomposition [30]. In stages V and VI, lignin is degraded, and in the next stage (VII) degradation prod-ucts from previous stages are decomposed. About 20% of residues remain after these processes, of which approx. 12 wt % is ash [31].
At work, Lina et al. [32] it was found that the share of basic and primary components of SCG are: hemicellulose 39 ± 2 wt % of dry SCG, cellulose (glucose) 12 ± 1 wt %, lig-nin 24 ± 2 wt %. In the coffee grounds used for research, the share of basic ingredients determined on the basis of TGA analysis is similar: hemicellulose 42.7 wt %, cellu-lose 15.5 wt %, lignin 18.2 wt %.
Thermal analysis using TGA was supplemented by analysis using DSC. Figure 2 compares DSC and DTG thermograms. The DSC thermogram shows the thermal transitions of the samples between -80 and 360°C. The
thermogram obtained for SCG showed glass transition and two events. For SCG the first event was observed with a peak at 89°C, and an associated enthalpy change of 195.6 J/g. This event is related to the melting transi-tion that occurs over a range of temperature due to the presence of impurities in the sample, the vaporization of water (indicating the presence of hydrophilic groups) and the crystalline nature of the materials [32]. The sec-ond event correspsec-onds to an exothermic transition and was observed at 300°C for SCG, respectively, with an associated enthalpy change of 17.2 J/g. This transition is related to the thermal depolymerization and branching of the samples, occurring at temperature ranges varying between 265 and 330°C [38]. These are probably changes in hemicellulose. These transformations were described by Line et al. [32] are similar to those determined in the tested coffee samples. In tested SCG samples, glass transi-tion was observed at a temperature of about -2°C.
FT-IR spectra of SCG showed a typical materials com-posed of polysaccharides and aromatic polymers (Fig. 3). Crystalline cellulose is an organic compound with a linear chain of several β (1→4) linked D-glucose units.
Strong bands are visible on SCG FT-IR spectra at 1457 cm-1,
1374 cm-1 and 868 cm-1 was assigned to the β-linkage of
cellulose. Based on the peak of 1457 cm-1 for cellulose, two
peaks showed the existence of lipids or several acids in
SCG. In the wave number range at 3000–3600 cm-1 a high
intensity band is visible OH stretching in carboxylic acid or various sugars and amine groups in caffeine or
pro-teins. Bands at 2920 cm-1 and 2849 cm-1 are attributed to
asymmetric and symmetric stretching of C–H bonds of methyl group in the caffeine molecule along with sug-ars [33]. These bands can also be associated with asym-metric and symasym-metric C–H stretching in lipids [34]. The
lipid is also associated with the band at 1736 cm-1, which
was attributed to carbonyl vibration (C=O) in aliphatic ester or triglycerides [35]. The spectrum FT-IR the band
at the 1510 cm-1 vibration is poorly visible. This
vibra-tion was due to C=C stretching of aromatic ring in lignin
or lipids [36]. Several peaks, 1245 cm-1 (C–O stretching),
1154 cm-1 (C–O–C asymmetric stretching) and 1027 cm-1
(C–O stretching), are attributed a great variety of mono-saccharides together with diverse acids molecules, such as chlorogenic acid, caffeic acid and coumaric acid [37].
Images obtained by scanning electron microscopy present that the SCG particles are composed of thin
0.1 0.0 -0.1 -0.2 -2°C -0.3 -0.4 -0.5 195.6 J/g -0.6 -0.7 -0.8 -0.9 Heat flow ,W/g Exo up 35°C 89°C 120°C 200°C -80 0 80 160 240 320 Temperature, °C 400 480 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Deriv ativ e mass, %/°C 345°C 314°C 300°C 17.2 J/g
Fig. 2. DSC and DTG thermograms of SCG
0.25 0.20 0.15 0.10 0.05 0.00 4000 3500 3000 2500 2000 1500 1000 500 718 809 868 1154 1245 1374 1457 1651 1736 1059 1027 1009 Absorbance, a.u. Wave number, cm-1 2849 2920 3273 3470
sheets – petals (Fig. 4). In her work, Lina et al. [32] also stated that SCG is material with very low porosity.
The particle size of the filler in isopropanol was ana-lyzed using a laser analyzer, and the results of particle size analysis are presented in Fig. 5. SCG particles have a size in the range 13–890 µm. The median particle size is 221 µm (std dev. 165 um). Two maxima were observed on the curve with particle sizes of about 77 µm and 344 µm.
Characteristics of VPF foam and its SCG composites
Addition of a filler causes a changes in the reaction kinetics, which was observed by analyzing the change
in foam growth time (tg) ( Fig. 6).
As a result of this analysis, it was found that by
add-ing 10 wt % of the filler, the tg value is clearly shortened,
which can be associated with an increase in the number of pores, because the filler particles can act as additional pore germs. The process of nucleating gas bubbles can be modified, among others by using fillers, e.g. in the form of nanoparticles, which will act as hetero zips [38]. Thanks to the hetero nucleation process leading to the forma-tion of gas bubbles, it is possible to obtain foam struc-tures with reduced pore sizes [39]. As a result of the
ana-lyzes it was found that used SCG filler contain approx. 3.6 wt % water. Water put into foam formulations signifi-cantly increases the demand for isocyanate. Therefore, it was analyzed how changes in the content of water added with the filler affect the change in the isocyanate index (INCO) of composites (Table 3).
T a b l e 3. Dependence of the isocyanate index on the amount of water added from SCG
Sample name VPF0 VPF10 VPF20 VPF30 VPF40 Added H20 wt % 0 0.36 0.72 1.08 1.44 Adjusted NCO 90 87 84 81 78
During the analysis of foams, Krebs and Hubel [40] found that a decrease in INCO results in a decrease in the degree of crosslinking of VEPUR foams, a decrease in porosity and a slight increase in the growth time of these foams. In order to clarify whether similar changes caused the addition of SCG, foam growth time was ana-lyzed (Fig. 6). Increasing the proportion of filler causes that the growth time of foams is longer, it also causes a decrease in INCO in foam formulations. By Krebs and Hubel should cause a slight increase in foam growth time, which was also observed in the tested foams. The reduction in growth time of VPF10 compared to foam without filler, may be due to the incorporation of min-erals from SCG. Most often they are various metal com-pounds that can act as catalysts. Saunders and Frish [41] ranked metal compounds according to their activity as catalysts, starting with the most active in the following sequence: Bi, Pb, Sn, Ti, Fe, Sb, U, Cd, Co, Th, Al, Hg, Zn, Ni, Ce, Mo, V, Cu, Mn, Zr.
The results presented in the paper Lin et al. [32] indicate that SCG contained large amounts of mineral compounds. Among the components mentioned in the analysis were significant amounts of iron, aluminum, manganese, zinc and cobalt compounds, which are catalysts for the reac-tion of OH and NCO groups.
In the foams presented in Fig. 7, a reduction in the pore size was observed, an increase in wall thickness 20 m 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Content, % 13 17 23 30 39 51 68 89 116 152 200 262 344 452 592 777 Diameter, m
Fig. 4. Micrograph by scanning electron microscopy of spent cof-fee grounds particles; magnification 500x
Fig. 5. SCG filler particle size distribution
120 116 112 108 104 100 0 10 20 30 40 50 Content of SCG, wt % Time of foam growth, s
Fig. 6. Influence of the filler additionon the change in growth time (tg) of produced foams
and increase in number of hole in the pore walls, these changes indicate a decrease in porosity in the analyzed materials along with an increase in the amount of filler introduced. These results are consistent with those pre-sented by Krebs and Hubel [40], determining the effect of INCO on porosity.
FT-IR spectra were performed to describe the chemical structure of VPF foam and its SCG composites. Figure 8 compares the spectra of foam and its composite contain-ing 40 wt % of SCG.
The spectra are representative of samples which are the subject of the article. In range of wave number 3300–
3500 cm-1 appear bands also from reaction substrates. Wave
number 3335 cm-1 and 3285 cm-1 originates from
stretch-ing vibration, symmetrical and asymmetrical, assigned to N-H bond. The band in the vicinity of wave number
3500 cm-1 is derived from the group –OH from water
or unbound polyols. Clearly outlining a range of wave
number 2867 cm-1 and 2970 cm-1 originates from
stretch-ing vibrations within groups –CH2 in the soft segments
formed from polyols [42, 43]. There was no band in the
vicinity of wave number 2270 cm-1 attributable to the
bind-ing of –NCO derived from unreacted isocyanate [42]. In all the analyzed samples were also observed bands derived
from bonds vibrations of C=O (1708 cm-1), C=C from
aro-matic ring (1597 cm-1) bending and deformation
vibra-tions derived from N–H bonds within HNC=O (1538 and
1511 cm-1), H
3C–C (1458 cm-1), –O–CH2 (1413 cm-1) and νasym
CO/sym within the group –NCO–O (1234 and 917 cm-1)
in –C–O–C– group [42]. In the range around 760 cm-1 the
band represents a C–H bond from the aromatic ring.
1 mm 1 mm 1 mm 1 mm a) b) c) d)
Fig. 7. SEM images of: a) VPF foam and its composites: b) 10 wt % of SCG, c) 30 wt % of SCG, d) 40 wt % of SCG
3500 4000 3000 2500 2000 1500 1000 500 Wave number, cm-1 0.4 0.2VPF40 0.0 0.2 0.40.0VPF0 Absorbance, a.u. 2973 2864 1310 1236 1083 1513 1536 1595 1713 1739 3296 3344 3335 3285 2973 2867 1713 16 04 1539 1507 13 10 1230 1080
Thermogravimetric analysis was also carried out for the foams, the results are illustrated in Fig. 9.
On the basis of the TG thermogram curves, the temper-ature of 2% mass loss (T2%), tempertemper-ature of 5% mass loss (T5%) and degradation residue at 600°C were determined.
However, on the basis of DTG curves, the temperature of the beginning and end of individual degradation stages
Ti and Ti+1, as well as the mass loss at the beginning and
end of each stage (mi and mi+1) were determined, which
allowed the determination of the change in mass during
Δmi = (mi+1 – mi). The maximum degradation rate (vmax) and
the temperature at which the maximum degradation rate
(Tmax) of each stage was reached were also determined.
The course of the analysis is schematically presented in Fig. 9, and the results of the analysis for all foams are pre-sented in Table 4. This table also indicates the tendency in change for each of the foam features.
The introduction of a filler causes a clear decrease in the 2% mass loss temperature associated with the loss of water and easily volatile substances contained in foams, and a T5% temperature often considered as the onset tem-perature of polyurethane degradation. The more filler in the biocomposite, the lower the T5%. The first stage of degradation of foam and biocomposites takes place in the range of 200 to 306 ± 6°C, the second stage of degradation ends at 369 ± 1°C, and the third ends at 455 ± 3°C. During the first stage of the decomposition process, the hard phase of foams degrade, probably most urethane bonds, and in composites additionally the degradation of hemi-cellulose. The mass loss at this stage decrease and change by approx. 9–14 wt %. The second stage of the process is associated with the degradation of the hard phase (urea bonds) and cellulose filler in composites. Mass loss at this stage may result from the distribution of urea bonds in the hard phase [44, 45]. At this stage, the degradation rate in composites increases and the mass loss increases and change by approx. 19–25 wt %.
The third degradation stage is related to the soft phase distribution of foams and the lignin in composites con-tained in the filler. In the third stage of degradation, the rate of degradation in composites significantly decreases with increasing amount of used filler. The mass loss in this stage decreases. After decomposition at 600°C, as the
0 20 40 60 80 100 Mass, % 1.5 1.0 0.5 0.0 Deriv ativ e mass, %/°C 680 600 520 440 360 280 200 120 40 Temperature, °C T2% T5% T max3 Tmax 2 Tmax 1 T1 T 2 T 3 T 4 U600 VPF0 VPF30 Cofee
Fig. 9. Comparison of TGA and DTG of coffee, VPF0 and VPF30 curves
T a b l e 4. The results of the analysis of TG and DTG curves of foams
Parameter/sample VPF0 VPF10 VPF20 VPF30 VPF40 T2%, °C 254 240 232 214 193 T5%, °C 280 273 273 270 266 T1, °C 200 200 200 200 200 m1, % 99.3 98.8 98.6 98.3 97.9 Tmax1, °C 306 302 295 292 285 vmax1, %/°C 0.30 0.30 0.22 0.22 0.25 T2, °C 313 312 303 299 304 m2, % 85.8 84.7 89.3 89.5 86.5 Δm1, % 13.8 14.1 9.3 8.8 11.4 Tmax2, °C 338 340 338 332 334 vmax2, %/°C 0.43 0.45 0.50 0.52 0.47 T3, °C 366 369 370 368 372 m3, % 66.8 64.1 64.4 63.6 63.0 Δm2, % 19.0 20.6 25.2 25.9 23.5 Tmax3, °C 413 413 416 415 419 vmax3, %/°C 1.23 1.28 1.25 1.23 1.14 T4, °C 453 453 456 456 458 m4, % 13.7 12.5 14.5 14.6 18.0 Δm3, % 53.1 51.6 49.9 49,0 45.7 U600, % 10.0 8.6 10.3 10.5 13.8
filler increases, the amount of ash remaining after degra-dation of the filler increases.
Trends in mass changes at individual degradation stages are shown in Fig. 10. Addition to 30 wt % SCG causes a decrease in mass loss in the first stage of hard phase degradation and an increase in mass loss in its sec-ond stage. Such changes indicate that in the first stage of hard phase degradation urethane bonds break down and in the second stage urea bonds. Probably the share of urea bonds increases with increasing amounts of water introduced with filler. Mass changes associated with hard phase distribution overlap those associated with SCG distribution. The reduction in mass loss in the stage associated with the distribution of the soft phase of poly-urethane is related with an increase in the filler content, which may be due to the large amount of ash remaining after SCG degradation.
Figure 11 presented DSC thermograms for unmodified foam, composite foams with SCGs, while Table 5 summa-rizes the results of the analysis.
The viscoelastic foams are made up of three phases: a soft phase, a phase composed of mixed flexible and rigid segments, and a hard phase [44]. During the first heating cycle, the occurrence was observed on thermograms of
VPF foams the glass transition temperature (Tg1) and
endo-thermic peaks associated with the change in order of mixed
flexible and rigid segments and a hard phase (Td, ΔHd). In
the tested foam and composites with coffee, the glass tran-sition temperature was observed in the soft phase at about -64 ± 1°C, the temperature associated with the change in order in the phase in the range 63–84°C and changes the enthalpy in the range 58–85 J/g. During the second heating cycle, three glass transition transformations were observed.
The first Tg1’ is associated with the vitrification of the soft
phase of foams, the second (Tg2) with the vitrification of the
mixture of flexible and rigid segments, and the third with
the vitrification of the hard phase (Tg3).
Thermal effects associated with transformations in composite foams overlap with thermal effects occurring in the filler. This makes the determination of character-istic transformations difficult and may increase the error of their determination. The addition of the filler does not
affect the change of Tg1’, the more filler the higher glass
transition temperature of the mixture of flexible and rigid
segments Tg2, which increases from -16 to -5°C.
The glass transition temperature of the hard phase VPF0 is around 39°C, the addition of filler causes a
sig- m1 m2 m3 70 60 50 40 30 20 10 0 m , w t % 0 10 20 30 40 SCG content, wt %
Fig. 10. Analysis of mass changes in subsequent stages of ther-mal degradation of VPF/SCG composites
-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 -64.8°C Tg1 -63.5°C Heat flow ,W/g Hm Tm 64.3 J/g 79.2°C 74.8 J/g 70.2°C -80 -40 0 40 80 120 160 200 Temperature, °C Exo up VPF0 VPF30 0.0 -0.1 -0.2 -0.3 -.0.4 -0.5 -0.6 -0.7 Exo up Heat flow ,W/g -80 -40 0 40 80 120 160 Temperature, °C 200 Tg1' -63.2°C Tg2 -11.8°C -62.5°C -15.0°C 61.4°C Tg3 38.6°C VPF0 VPF30
Fig. 11. Comparison of DSC thermograms of unmodified foam and composite foams with SCGs: a) first cycle of heating, b) second cycle of heating
a) b)
T a b l e 5. Results of the DSC analysis of examined materials
Sample Tg1 °C °CTd ΔHJ/gd T°Cg1’ T°Cg2 T°Cg3 VPF0 -62 71 85.3 -62 -15 39 VPF10 -64 84 67.5 -63 -16 57 VPF20 -64 63 58.0 -63 -13 43 VPF30 -65 79 73.2 -63 -12 60 VPF40 -65 72 73.2 -64 -5 58
nificant increase in Tg3 to a temperature in the range 43–60°C. Changes in the micro- and macrostructure of foams caused by the addition of the filler affects changes in their properties. Table 6 summarizes the results of the apparent density (D), calculations of the theoretical value
of apparent density (Dt), and differences between
theo-retical apparent density and apparent density (Dt - D).
T a b l e 6. Apparent density of tested materials
Materials Apparent density
D, kg/m³ Theoretical apparent density Dt, kg/m³ Dt - D kg/m³ VPF0 40.90 ± 0.35 – – VPF10 41.09 ± 0.26 73.5 32.4 VPF20 43.14 ± 0.52 107.5 64.4 VPF30 50.90 ± 1.41 145.1 94.2 VPF40 54.18 ± 2.36 178.5 124.3 The apparent density increases with the amount of used filler. To explain the reason for these changes, the theoretical density of foams was calculated using the mix-ture rule for calculations. It was assumed that the density of the matrix in composite foams is equal to the density
of VPF0 and the density of coffee is equal 365 kg/m3 [46].
The theoretical apparent density of composite foams was higher than the apparent density, the difference between the density values for composite foams increased as the amount of filler increased. These results indicate that increased porosity of composite caused by the reaction of water introduced with the filler and NCO groups of isocyanate.
Low compression set of VPF is important, e.g. during transport, but also during use. Table 7 summarizes the results of determining the value of compression set for all tested samples. The VPF0 foams were significantly deformed as a result of the test, especially in the case of compression by 75% and 90% of the original height. After addition the filler, all foams have permanent deforma-tions with a value below the permissible limit of 10%.
Table 8 contains the values determined during the fourth foam compression cycle: comfort factor (SAG) and hardness parameter (F40%). Increasing the proportion of filler in foams causes that the comfort factor increases
and the foam hardness decreases. Such changes in these properties are probably the result of a decrease in INCO due to the introduction of additional water into the reac-tion mixture. The hardness of the foams is not affected by the hardness of the filler particles.
T a b l e 8. Properties determined in the compression test of materials Materials F40% Pa SAG VPF0 3.3 ± 0.3 2.1 ± 0.1 VPF10 2.3 ± 0.2 2.1 ± 0.1 VPF20 1.9 ± 0.2 2.2 ± 0.1 VPF30 1.8 ± 0.2 2.6 ± 0.2 VPF40 2.0 ± 0.2 2.7 ± 0.2
Analysis of resistance to flame action in a cone calo-rimeter allowed to establish that the use of 20 wt % coffee
particles reduces the maximum heat rate release (HRRmax)
by approx. 12% (Fig. 12). The total heat release (THR) of the composite decreases slightly, but clearly decreases by about 28% by mass loss rate (MLR). Such non-flammabil-ity properties of foam with the addition of 20 wt % SCG may be the result of an increase in the amount of carbon in the combusted material, as indicated by a significant amount of ash and SCG.
CONCLUSIONS
The spent coffee grounds used to make composites are built like typical plant fillers made of hemicellulose, cellulose and lignin as well as secondary ingredients. The SCG contains: hemicellulose 42.7 wt %, cellulose 15.5 wt %, lignin 18.2 wt %, about 3.6 wt % water and 1.1% of secondary ingredients. About 20% of the mass remains after degradation at 600°C. The chemical struc-ture of the coffee particles was confirmed by analysis using FT-IR and DSC. Coffee particles have sizes in the range 13–890 µm and particle size of the filler is approx. T a b l e 7. Results of measurements of permanent deformation
after compression of foams
Sample The compression set (CS)
22 h, 70°C, 50% 22 h, 70°C, 75% 22 h, 70°C, 90% VPF0 2.1 ± 0.7 10.0 ± 0.9 85.2 ± 2.7 VPF10 1.1 ± 0.7 1.8 ± 1.1 2.8 ± 1.1 VPF20 2.7±0.6 3.8±1.5 4.2±1.2 VPF30 1.4 ± 0.8 2.3 ±0.5 2.8 ± 0.8 VPF40 2.9 ± 0.4 3.2 ± 1.0 3.1 ± 0.4 140 120 100 80 60 40 20 0 0 50 100 150 200 250 Time, s Heat release rate, kW/m 2 VPF0 VPF20
Fig. 12. Change in heat release rate as a function of time VPF0 and VPF20
221 µm, they are characterized by low porosity. Addition of 10 wt % SCG filler into VPF foam reduces foam growth time by approx. 11%. The reason for this change in the reaction kinetics is the use of filler with mineral com-ponents – many metal compounds that act as catalysts for the reaction of -OH and -NCO groups. Increasing the amount of filler increases the amount of water in the VPF reaction. This results in a lowering of the index of isocya-nate in composites and, as a result, longer foam growth time. A larger amount of water also reduces the size of the pores, increases the wall thickness and of the ovals in the walls joining the pores.
These changes indicate a decrease in porosity in the analyzed materials resulting from a decrease in the index of isocyanate. The addition of a filler caused a change in the thermal characteristics and thermal degradation of composites. In foams with SCG filler, the temperature at which the release of volatile products begins and the degradation process begins decreases. However, the rate of degradation in the first stage of degradation associated with the hard phase degradation decreases, but the rate of degradation of the second phase of the hard phase deg-radation increases. Changes in the hard phase decompo-sition process result from changes caused by presence of additional water into the VPF formation environment.
The introduction of a filler reduces the degradation rate of the soft phase polyurethane foams.
The addition of a filler affected the change in apparent density of foams resulting from use a higher density filler and slight increase in foam porosity due to the
forma-tion of more CO2 as a result of the reaction of isocyanate
with water. Compression set of foams after compression by 75% and 90% after the addition of SCG are extremely advantageously reduced.
The hardness of composites also decreased, which is significantly influenced by the change in the stiffness of joints between the pores in which the filler particles are embedded.
The comfort factor increased after adding SCG, which may be due to a decrease in the isocyanate index. The change in the isocyanate index is the result of introduc-ing more water together with the filler. The flame retar-dancy of foams is the result of introducing more carbon with the filler. As a result, the heat release rate and smoke release rate are reduced.
REFERENCES
[1] Figueroa G., Homann T., Rawel H.: Austin Food
Sciences 2016, 1, 1014.
[2] Campos-Vega R., Loarca-Pina G., Vergara-Castaneda H.A. et al.: Trends in Food Science & Technology 2015,
45, 24.
https://doi.org/10.1016/j.tifs.2015.04.012
[3] Cruz R., Cardoso M.M., Fernandes L. et al.: Journal of
Agricultural and Food Chemistry 2012, 60, 7777.
https://doi.org/10.1021/jf3018854
[4] Murthy P.S., Naidu M.M.: Resources, Conservation and
Recycling 2012, 66, 45.
https://doi.org/10.1016/j.resconrec.2012.06.005
[5] Caetano N.S., Silva V.F., Mata T.M.: Italian Association
of Chemical Engineering 2012, 26, 267.
https://doi.org/10.3303/CET1226045
[6] Park J., Kim B., Son J. et al.: Bioresource Technology
2018, 249, 494.
https://doi.org/10.1016/j.biortech.2017.10.048
[7] Mussatto S.I., Carneiro L.M., Silva J. et al.: Carbohydrate
Polymers 2011, 83, 368.
https://doi.org/10.1016/j.carbpol.2010.07.063
[8] Fiol N., Escudero C., Villaescusa I.: Separation Science
and Technology 2008, 43, 582.
https://doi.org/10.1080/01496390701812418
[9] Roussos S., de los Angeles Aquiahuatl M., del Refugio Trejo-Hernandez M. et al.: Applied Microbiology and
Biotechnology 1995, 42, 756.
https://doi.org/10.1007/BF00171958
[10] Ballesteros L.F., Teixeira J.A., Mussatto S.L.: Food and
Bioprocess Technology Provides 2014, 7, 3493.
https://doi.org/10.1007/s11947-014-1349-z
[11] Farach A., Donangelo C.M.: Brazilian Journal of Plant
Physiology 2006, 18, 1677.
https://doi.org/10.1590/S1677-04202006000100003 [12] Mendes L., Mendes R.F., Tonoli G.H.D. et al.: Key
Engineering Materials 2012, 517, 556.
https://doi.org/10.4028/www.scientific.net/ KEM.517.556
[13] Esquivel P., Jiménez V.M.: Food Research International
2012, 46, 488.
https://doi.org/10.1016/j.foodres.2011.05.028
[14] Mussatto S.I., Ballesteros L.F., Martins S. et al.:
Separation and Purification Technology 2011, 83, 173.
https://doi.org/10.1016/j.seppur.2011.09.036
[15] Arya M., Rao L.J.M.: Critical Reviews in Food Science
and Nutrition 2007, 47, 51.
https://doi.org/10.1080/10408390600550315
[16] Burniol-Figols A., Ceniana K., Skiadas I.V. et al.:
Biochemical Engineering Journal 2016, 116, 54.
https://doi.org/10.1016/j.bej.2016.04.025
[17] Wang H.M.D., Cheng Y.S., Huang C.H. et al.: Applied
Biochemistry and Biotechnology 2016, 180, 753.
https://doi.org/10.1007/s12010-016-2130-8
[18] Ryszkowska J.: „Materiały poliuretanowe wytwa-rzane z zastosowaniem surowców odnawialnych”, Oficyna Wydawnicza Politechniki Warszawskiej, Warszawa 2019.
[19] Kintrup S., Treboux J.P., Mispreuve H.: Proceedings of the Polyurethanes Conference, Boston, Massachusetts, 8–10 October 2000, p. 91.
[20] Hager S., Skorpenske R., Triouleyre S. et al.: Journal of
Cellular Plastics 2001, 37, 377.
https://doi.org/10.1106/TQ0X-NBBV-RPD6-VWHH [21] Krebs M., Hubel R.: “Highly Sophisticated Cell Opener
for Viscoelastic Foam”, Evonik Industries AG Nutrition & Care Comfort & Insulation, Essen Germany.
ht t ps://w w w.pu-addit ives.com/.../High ly _ Sophisticate_Cell_Opener_for_ Viscoelastic_Foam. pdf (10.04.2016). https://www.pu-additives.com/product/pu-additi-ves/downloads/highly_sophisticated_cell_opener_ for_viscoelastic_foam_paper.pdf
[22] Yao F., Wu Q., Lei Y. et al.: Polymer Degradation and
Stability 2008, 93, 90.
https://doi.org/10.1016/j.polymdegradstab.2007.10.012 [23] Ramiah M.V.: Journal of Applied Polymer Science 1970,
14, 1323.
https://doi.org/10.1002/app.1970.070140518
[24] Guimarães J.L., Frollini E., da Silva C.G. et al.:
Industrial Crops and Products 2009, 30, 407.
https://doi.org/10.1016/j.indcrop.2009.07.013
[25] Rowell R.M.: “Properties and performance of natu-ral-fibre composites” in “Woodhead Publishing Series in Composites Science and Engineering” 2008, pp. 3–66.
[26] Faruk O., Bledzki A.K., Fink H.P. et al.: Progress in
Polymer Science 2012, 37, 1552.
https://doi.org/10.1016/j.progpolymsci.2012.04.003 [27] Gourier C., Le Duigou A., Bourmaud A. et al.:
Composites Part A: Applied Science and Manufacturing
2014, 64, 159.
https://doi.org/10.1016/j.compositesa.2014.05.006 [28] Väisänen T., Haapala A., Lappalainen R. et al.: Waste
Management 2016, 54, 62.
https://doi.org/10.1016/j.wasman.2016.04.037
[29] Chun Y., Gun Ko Y., Do T. et al.: Colloids and Surfaces
A: Physicochemical and Engineering Aspects 2019, 562,
392.
https://doi.org/10.1016/j.colsurfa.2018.11.005 [30] Yang H., Yan R., Chen H. et al.: Fuel 2007, 86, 1781.
https://doi.org/10.1016/j.fuel.2006.12.013
[31] Ribeiro J.P., Vicente E.D., Nunes M.I. et al.:
Environmental Science and Pollution Research 2017, 24,
15270.
https://doi.org/10.1007/s11356-017-9134-y
[32] Ballesteros L.F., Teixeira J.A., Mussatto S.I.: Food and
Bioprocess Technology 2014, 7, 3493.
https://doi.org/10.1007/s11947-014-1349-z
[33] Kemsley E.K., Ruault S., Wilson R.H.: Food Chemistry
1995, 54, 32.
https://doi.org/10.1016/0308-8146(95)00030-M
[34] Craig A.P., Franca A.S., Oliveira L.S. et al.: Talanta
2014, 128, 393.
https://doi.org/10.1016/j.talanta.2014.05.001
[35] Lyman D.J., Benck R., Dell S. et al.: Journal of
Agricultural Food Chemistry 2003, 51, 3268.
https://doi.org/10.1021/jf0209793
[36] Pandey K.K.: Journal of Applied Polymer Science 1999,
71, 1969.
h t t p s : / / d o i . o r g / 1 0 . 1 0 0 2 / ( S I C I ) 1 0 9 7 -4628(19990321)71:12<1969::AID-APP6>3.0.CO;2-D [37] Clifford M.N., Kirkpatrick J., Kuhnert N. et al.: Food
Chemistry 2008, 106, 379.
https://doi.org/10.1016/j.foodchem.2007.05.081
[38] Saunders J.H., Hansen R.H.: “Plastic Foams”, Marcel Dekker, New York, 1972, pp. 23–108.
[39] Saunders J.H., Hansen R.H.: “The Mechanism of Foam Formation” in “Plastic Foams, Part I”, Marcel Dekker: New York, 1972, pp. 23–108.
[40] Krebs M., Hubel R.: “The Adjustment of Physical Properties of Viscoelastic Foam – the Role of Different Raw Materials”, American Chemistry Council, 2016. https://www.pu-additives.com/product/pu-additi-ves/downloads/adjustments-of-physical-properties. pdf.
[41] Saunders J.H., Frisch K.C.: “Polyurethanes: Chemistry and Technology. Chemistry”, RE Krieger, 1987.
[42] Hatchett D.W., Kinyanju J.M., Sapochak L.: Journal of
Cellular Plastics 2007, 43, 183.
http://dx.doi.org/10.1177/0021955X07076665
[43] Zieleniewska M., Szczepkowski L., Krzyżowska M.
et al.: Polimery 2016, 61, 807.
http://dx.doi.org/10.14314/polimery.2016.807
[44] Bernardini J., Angullesi I., Coltelli M.B. et al.: Polymer
International 2015, 64, 1235.
https://doi.org/10.1002/pi.4905
[45] Demirci F., Yildirim K., Kocer H.B.: Journal of Applied
Polymer Science 2018, 135, 45914.
https://doi.org/10.1002/app.45914
[46] Ammerlaan T., Barrière V., Genest-Richard P. et al.: “Tales of a Forgotten Bioresource: The Recycling of Spent Coffee Grounds”, Department of Bioresource Engineering, McGill University Montréal, Canada, 23.04.2012.
https://www.mcgill.ca/bioeng/files/bioeng/thomas_ violette_sandra_pascal_2012.pdf.
[47] Schnabel W.: “Polymer degradation. Principles and applications”, Akademie-Verlag, Berlin 1981.
